[MASTER] multiple failures on evolution-data-server-2.28 and gconf

[cbinusa]

using latest karmic snapshot.

ProblemType: Crash
Architecture: amd64
CrashCounter: 1
Date: Tue Aug 11 20:44:10 2009
DistroRelease: Ubuntu 9.10
ExecutablePath: /usr/lib/evolution/evolution-data-server-2.28
NonfreeKernelModules: nvidia
Package: evolution-data-server 2.27.90-0ubuntu1
ProcCmdline: /usr/lib/evolution/evolution-data-server-2.28 --oaf-activate-iid=OAFIID:GNOME_Evolution_DataServer_CalFactory:1.2 --oaf-ior-fd=26
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
 PATH=(custom, user)
ProcVersionSignature: Ubuntu 2.6.31-5.24-generic
SegvAnalysis:
 Segfault happened at: 0x7f9ef3a1e861: mov (%rdi),%eax
 PC (0x7f9ef3a1e861) ok
 source "(%rdi)" (0x00000000) not located in a known VMA region (needed readable region)!
 destination "%eax" ok
SegvReason: reading NULL VMA
Signal: 11
SourcePackage: evolution-data-server
StacktraceTop:
 ?? () from /lib/libdbus-1.so.3
 ?? () from /lib/libdbus-1.so.3
 ?? () from /lib/libdbus-1.so.3
 ?? () from /lib/libdbus-1.so.3
 ?? () from /lib/libdbus-1.so.3
Title: evolution-data-server-2.28 crashed with SIGSEGV
Uname: Linux 2.6.31-5-generic x86_64
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

[Apport retracing service (apport)]

StacktraceTop:_dbus_watch_unref (watch=0x0) at dbus-watch.c:128
free_watches (transport=0x2408b60)
socket_disconnect (transport=0x0)
_dbus_transport_disconnect (transport=0x2408b60)
_dbus_transport_queue_messages (transport=0x2408b60)

[Scott James Remnant (Canonical) (canonical-scott)]

Please don't just reassign bugs to libdbus without first verifying that they are dbus bugs; this looks like an evolution-data-server bug (freeing an already free watch); it could also be a libdbus-glib bug, gvfs bug, etc.

[Pedro Villavicencio (pedro)]

Thanks for your bug report. This bug has been reported to the developers of the software. You can track it and make comments here: http://bugzilla.gnome.org/show_bug.cgi?id=593017

[C de-Avillez (hggdh2)]

Well. The basic issue here seems to be that gconf is *not* thread-safe, and Evolution/EDS is heavily threaded. Looking at all the EDS bugs, I can see many of them where we have more than one thread running gconf.

I am not yet sure why we are getting these errors mostly on Karmic; we should have them pretty much since Gutsy/Hardy.

[C de-Avillez (hggdh2)]

An interesting tidbit: most -- if not all -- of the bugs I dupped here have, as the ProCCmdLine the following:

/usr/lib/evolution/evolution-data-server-2.28 --oaf-activate-iid=OAFIID:GNOME_Evolution_DataServer_CalFactory:1.2 --oaf-ior-fd=27

All of them have at least two threads with gconf calls in them, and all of them got the signal in one such thread.